[
Lists Home |
Date Index |
Thread Index
]
Paul Prescod wrote:
>> limited SGML support. Moreover, SGML didn't have the predefined entities
>> for < and &, so CDATA must have been the only standard alternative.
>No, you could use numeric entities. Or define "<" and "&" in the DTD or
>internal subset.
I just meant that these were knocked together by the user, not something
always included in the syntax. My point is that there's an annoying overlap
between CDATA and the predefined entities. If people really use CDATA, then
fine. But in six years I never needed them myself, so wonder if they belong
to the SGML heritage along with notations, ANY and conditional sections.
>May Goldfarb save us from those who want to protect us from the syntax
>by making the syntax so unbearable that we need protection!
Good point.
Gustaf
|
